[TensorRT] Add TensorRT target and export qwen3-06b model

README.md

@@ -145,20 +145,31 @@ See the [CLI Reference](https://onnxruntime.github.io/mobius/cli_reference.html)
 
 ## Architecture
 
-```
-HuggingFace Hub
-       │
-       ▼
- ArchitectureConfig ◄── from_transformers() / from_diffusers()
-       │
-       ▼
- Model Module ◄── Reusable Components (Attention, MLP, RMSNorm, RoPE, …)
-       │
-       ▼
- Task ◄── CausalLMTask, VisionLanguageTask, VAETask, DenoisingTask, …
-       │
-       ▼
- ONNX Model ◄── preprocess_weights() + apply_weights()
+```mermaid
+flowchart TD
+    Sources["Model sources<br/>Transformers · Diffusers · GGUF · NeMo"]
+    Config["ArchitectureConfig<br/>Normalizes source configuration"]
+    Registry["Registry<br/>Selects the model class and task"]
+    Components["Reusable components<br/>Attention · MLP · Norm · RoPE · MoE · Vision · Audio"]
+    Models["Model modules<br/>Compose components into architectures"]
+    Tasks["Tasks<br/>Define ONNX inputs, outputs, caches, and model splits"]
+    Graph["ONNX graph construction<br/>onnxscript.nn + onnx_ir"]
+    Optimize["EP-aware optimization<br/>Cleanup · Fusion · Lowering · Folding"]
+    Weights["Weight pipeline<br/>Download · Rename · Transform · Cast · Apply"]
+    Package["ModelPackage<br/>One or more deployable ONNX models"]
+    Runtime["ONNX Runtime / ONNX Runtime GenAI"]
+
+    Sources --> Config
+    Config --> Registry
+    Registry --> Models
+    Registry --> Tasks
+    Components --> Models
+    Models --> Tasks
+    Tasks --> Graph
+    Graph --> Optimize
+    Optimize --> Weights
+    Weights --> Package
+    Package --> Runtime
 ```
 
 The package is organised into four layers:
@@ -171,6 +182,37 @@ The package is organised into four layers:
 
 See the [design document](https://onnxruntime.github.io/mobius/design.html) for details.
 
+### Repository organization
+
+```mermaid
+flowchart LR
+    Root["src/mobius/"]
+    Root --> API["Public API and build orchestration<br/>__init__.py · _builder.py · _model_package.py"]
+    Root --> Configs["_configs/<br/>Normalized architecture configuration"]
+    Root --> Components["components/<br/>Reusable ONNX building blocks"]
+    Root --> Models["models/<br/>Architecture implementations"]
+    Root --> Tasks["tasks/<br/>Graph I/O contracts"]
+    Root --> Registry["_registry.py<br/>Model and task lookup"]
+    Root --> Optimizations["_optimizations.py · rewrite_rules/ · _passes/<br/>Graph optimization"]
+    Root --> Integrations["integrations/<br/>Transformers · Diffusers · GGUF · NeMo · ORT GenAI"]
+
+    Configs --> Models
+    Components --> Models
+    Registry --> Models
+    Registry --> Tasks
+    Models --> Tasks
+    Tasks --> API
+    Integrations --> API
+    API --> Optimizations
+```
+
+Supporting directories:
+
+- `tests/` contains graph-construction, integration, parity, generation, and runtime tests.
+- `src/mobius/**/*_test.py` contains unit tests co-located with their implementation.
+- `examples/` demonstrates text, multimodal, speech, and diffusion workflows.
+- `docs/` contains user guides, design documentation, and API reference material.
+
 ## Development
 
 ```bash

docs/execution_providers.md

@@ -145,6 +145,11 @@ EpCapabilities(name="webgpu",  gqa_dtypes={FLOAT, FLOAT16},
 EpCapabilities(name="trt-rtx", gqa_dtypes={FLOAT16, BFLOAT16},
                supports_skip_layer_norm=False, enable_graph_capture=True,
                provider_options={"enable_cuda_graph": "1"})
+EpCapabilities(name="tensorrt",
+               static_cache_layout="heads_first",
+               supports_attention_nonpad_kv_seqlen=False,
+               gqa_dtypes=frozenset(), qkv_pack_dtypes=frozenset(),
+               supports_skip_layer_norm=False, supports_matmul_nbits=False)
 EpCapabilities(name="onnx-standard",
                gqa_dtypes=frozenset(), qkv_pack_dtypes=frozenset(),
                supports_fused_rope=False,  # not used by default for this EP, since it does not enable GQA fusion
@@ -153,6 +158,18 @@ EpCapabilities(name="onnx-standard",
                supports_packed_multi_head_attention=False)
 ```
 
+Standalone `tensorrt` is separate from ORT's `trt-rtx` provider. Static-cache
+exports use `[B, kv_heads, capacity, head_dim]` caches and an explicit additive
+attention bias, with `is_causal=0`. Query cache slots are
+`write_indices + arange(query_length)`; allowed keys must also be below
+`nonpad_kv_seqlen`. The length remains a graph input used by the bias, but is
+omitted from native Attention input #6: the tested TensorRT 11.3 path ignores
+that input and otherwise applies incorrect top-left causality during decode.
+Default/ORT masking behavior is unchanged. No extra `static-cache-bias` flag
+is required for standalone TensorRT. Other model backbones must supply a full
+static-cache bias or export fails explicitly instead of emitting maskless
+attention on this provider.
+
 Out-of-tree EPs can register at runtime via `register_ep()`:
 
 ```python

docs/index.md

@@ -52,4 +52,5 @@ ai-model-support-strategy
 :caption: Research
 
 research/testing-strategy-analysis
+research/tensorrt-static-cache-debugging
 ```